Deepfakes have raised significant concerns due to their potential to spread false information and compromise the integrity of digital media. Current deepfake detection models often struggle to generalize across a diverse range of deepfake generation techniques and video content. In this work, we propose a Generative Convolutional Vision Transformer (GenConViT) for deepfake video detection. Our model combines ConvNeXt and Swin Transformer models for feature extraction, and it utilizes an Autoencoder and Variational Autoencoder to learn from latent data distributions. By learning from the visual artifacts and latent data distribution, GenConViT achieves an improved performance in detecting a wide range of deepfake videos. The model is trained and evaluated on DFDC, FF++, TM, DeepfakeTIMIT, and Celeb-DF (v2) datasets. The proposed GenConViT model demonstrates strong performance in deepfake video detection, achieving high accuracy across the tested datasets. While our model shows promising results in deepfake video detection by leveraging visual and latent features, we demonstrate that further work is needed to improve its generalizability when encountering out-of-distribution data. Our model provides an effective solution for identifying a wide range of fake videos while preserving the integrity of media.
With the development of ubiquitous technologies that support the digitization of money, research is needed on how individuals’ private life practices are affected by new technological financial systems and how cash-based practices can inform their design. In this article, we report the cash-based monetary practices of one Ethiopian rural community and identify their implications for the design of new financial technology. Particularly, we focus on addressing the question, What characteristic features should go into the design of mobile money application(s) to embody a rural Ethiopian community’s money practices in social (marriage and death) and religious contexts? Primary data on everyday practices of the community were collected. Analysis of our data reveals that new financial technology design should support lived experiences such as embedded social meaning, segregated and aggregate money control, restricted money use, identity extension and hiding, refusal and acceptance of donations, disclosed and secret money practices, and assigning aesthetics to money. The article concludes by discussing possible ways of mapping these concepts into financial system design, thus contributing toward the development of cashless transactions and a cashless society.
The objective of this study was to develop information mining methodology for drought modeling and predictions using historical records of climate, satellite, environmental, and oceanic data. The classification and regression tree (CART) approach was used for extracting drought episodes at different time-lag prediction intervals. Using the CART approach, a number of successful model trees were constructed, which can easily be interpreted and used by decision makers in their drought management decisions. The regression rules produced by CART were found to have correlation coefficients from 0.71–0.95 in rules-alone modeling. The accuracies of the models were found to be higher in the instance and rules model (0.77–0.96) compared to the rules-alone model. From the experimental analysis, it was concluded that different combinations of the nearest neighbor and committee models significantly increase the performances of CART drought models. For more robust results from the developed methodology, it is recommended that future research focus on selecting relevant attributes for slow-onset drought episode identification and prediction.
The emergence of the World Wide Web facilitates the growth of user-generated texts in less-resourced languages. Sentiment analysis of these texts may serve as a key performance indicator of the quality of services delivered by companies and government institutions. The presence of user-generated texts is an opportunity for assisting managers and policy-makers. These texts are used to improve performance and increase the level of customers’ satisfaction. Because of this potential, sentiment analysis has been widely researched in the past few years. A plethora of approaches and tools have been developed—albeit predominantly for well-resourced languages such as English. Resources for less-resourced languages such as, in this paper, Amharic, are much less developed. As a result, it requires cost-effective approaches and massive amounts of annotated training data, calling for different approaches to be applied. This research investigates the performance of a combination of heterogeneous machine learning algorithms (base learners such as SVM, RF, and NB). These models in the framework are fused by a meta-learner (in this case, logistic regression) for Amharic sentiment classification. An annotated corpus is provided for evaluation of the classification framework. The proposed stacked approach applying SMOTE on TF-IDF characters (1,7) grams features has achieved an accuracy of 90%. The overall results of the meta-learner (i.e., stack ensemble) have revealed performance rise over the base learners with TF-IDF character n-grams.
The rapid advancement of deep learning models that can generate and synthesis hyper-realistic videos known as Deepfakes and their ease of access to the general public have raised concern from all concerned bodies to their possible malicious intent use. Deep learning techniques can now generate faces, swap faces between two subjects in a video, alter facial expressions, change gender, and alter facial features, to list a few. These powerful video manipulation methods have potential use in many fields. However, they also pose a looming threat to everyone if used for harmful purposes such as identity theft, phishing, and scam. In this work, we propose a Convolutional Vision Transformer for the detection of Deepfakes. The Convolutional Vision Transformer has two components: Convolutional Neural Network (CNN) and Vision Transformer (ViT). The CNN extracts learnable features while the ViT takes in the learned features as input and categorizes them using an attention mechanism. We trained our model on the DeepFake Detection Challenge Dataset (DFDC) and have achieved 91.5 percent accuracy, an AUC value of 0.91, and a loss value of 0.32. Our contribution is that we have added a CNN module to the ViT architecture and have achieved a competitive result on the DFDC dataset.
2 March 2023 in International Journal on Digital Libraries